Intelligent thrust control plan for aero-engine

Abstract

The invention discloses an aero-engine intelligent thrust control plan for a throttle lever and thrust high-linearity mapping demand, and in the thrust control plan, steady-state operation data is obtained by operating an engine component-level model; constructing a first intermediate / slow thrust estimator to provide respective first estimates of the intermediate thrust and the slow thrust based on the flight parameters; constructing a decline estimator to estimate thrust deviation caused by performance decline of the engine, and compensating the first estimated value to obtain corrected estimated values of the middle thrust and the slow thrust; constructing a second intermediate / slow thrust estimator, and performing online updating according to the corrected estimated values of the intermediate and slow thrust to achieve the purpose of adapting to performance degradation of the engine and obtain degradation self-adaptive intermediate and slow thrust estimation results; and constructing a reference thrust estimator, and determining the reference thrust by performing linear interpolation on the angle of the throttle lever by using the output result of the second intermediate / slow vehicle thrust estimator.

Description

Technical Field

[0001] This invention relates to the field of aero-engine control technology, and proposes an intelligent thrust control scheme for aero-engines that addresses the requirement of high linear mapping between throttle lever and thrust. Background Technology

[0002] The control plan for an aero-engine is the "way" that the control system follows. It encompasses the rules governing the changes in all controlled parameters during engine control operations, guiding the engine's operation under various conditions. However, the wide engine envelope, varied mission profiles, and numerous coupled control variables present significant challenges to control plan design.

[0003] One of the key functions of an aero-engine is to provide thrust to the aircraft. As a crucial performance indicator, thrust has long been central to engine design, control, and optimization. Fifth-generation fighter jets, in particular, prioritize improving thrust control quality, achieving thrust vectoring control, and integrated flight / engine control as core development requirements for their engine control systems. Unfortunately, because engine thrust cannot be directly measured during flight, neither the pilot nor the control system can monitor the engine's thrust status in real time. To meet the thrust control requirements of fighter jet combat missions, in practice, high-pressure or low-pressure rotor speeds, which are strongly correlated with thrust, are typically chosen as alternative control targets to construct speed control plans, thereby achieving indirect thrust control. However, under this speed control architecture, the pilot directly controls the rotor speed by manipulating the throttle, but directly experiences changes in engine thrust. Furthermore, within the full thrust envelope, the relationship between rotor speed and engine thrust typically exhibits strong nonlinear characteristics, and this relationship continues to change as the engine's service life increases. Therefore, it is difficult for pilots to achieve precise thrust adjustment through throttle operation. In practical applications, the following problems exist: (1) there is a non-preset dead zone for thrust change near engine idle and intermediate states; (2) the linearity between throttle lever angle and thrust is poor.

[0004] To meet the thrust requirements of different combat missions, pilots must be extremely familiar with the correspondence between throttle lever angle and thrust in order to achieve precise throttle control to a certain extent. Furthermore, pilots must constantly assess whether the current thrust changes meet mission expectations during throttle operation. If the expected control effect is not achieved, additional throttle adjustments must be made. Undoubtedly, frequent throttle adjustments increase the pilot's workload and may even affect the effective execution of combat missions.

[0005] Therefore, in order to alleviate the burden of throttle operation on pilots and ensure the high-quality execution of combat missions, it is necessary to construct an intelligent thrust control scheme for aero-engines that meets the requirements of high linear mapping between throttle lever and thrust, given the current situation that engine thrust cannot be directly measured during flight and the limitations of existing speed control schemes.

[0006] The information disclosed in the background section is only intended to enhance the understanding of the background of the present invention, and therefore may contain information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0007] To address the shortcomings or defects of the existing technology, an intelligent thrust control scheme for aero-engines is provided, which addresses the requirement for high linear mapping between throttle lever and thrust. This scheme can effectively reduce the non-preset dead zone of thrust changes and improve the linearity between throttle lever and thrust throughout the entire lifespan.

[0008] The objective of this invention is achieved through the following technical solutions.

[0009] A plan for intelligent thrust control of aero-engines, addressing the requirement for high linear mapping between throttle lever and thrust, includes:

[0010] S100: Steady-state operating data is obtained by running an engine component-level model, the steady-state operating data including flight parameters and engine air path parameters;

[0011] S200: Construct a first intermediate / idle thrust estimator to provide a first estimate of intermediate thrust and a first estimate of idle thrust based on flight parameters;

[0012] S300: Construct a decay estimator to estimate the thrust deviation caused by engine performance degradation based on flight parameters and engine air path parameters. The thrust deviation includes the thrust deviation of the first estimated intermediate thrust and the thrust deviation of the first estimated idle thrust.

[0013] S400: Sum the first estimate of intermediate thrust and the thrust deviation of the first estimate of intermediate thrust to calculate the corrected estimate of intermediate thrust; Sum the first estimate of idle thrust and the thrust deviation of the first estimate of idle thrust to calculate the corrected estimate of idle thrust.

[0014] S500: Construct a second intermediate / idle thrust estimator with the same structure as the first intermediate / idle thrust estimator. The second intermediate / idle thrust estimator is used to update online based on the corrected estimate of intermediate thrust and the corrected estimate of idle thrust, and finally provides a second estimate of intermediate thrust and a second estimate of idle thrust based on flight parameters.

[0015] S600: Construct a reference thrust estimator, using the second estimate of intermediate thrust and the second estimate of idle thrust, and determine the reference thrust by linear interpolation of the throttle lever angle.

[0016] In the intelligent thrust control program for aero-engines, which addresses the requirement for a high linear mapping between throttle lever and thrust, flight parameters include altitude (H), Mach number (M), and throttle lever (PLA); engine airflow parameters include low-pressure rotor speed (n). L High-voltage rotor speed n H Total temperature T at the fan inlet t2 Low-pressure turbine outlet total temperature T t6 Total pressure at the fan inlet p t2 .

[0017] In the intelligent thrust control program for aero-engines that addresses the requirement of high linear mapping between throttle lever and thrust, the first intermediate / idle thrust estimator is:

[0018] ,

[0019] in, This represents the input-output mapping relationship of the first intermediate / slow train thrust estimator. and These are the input and output vectors of the first intermediate / slow train thrust estimator, respectively.

[0020] In the intelligent thrust control program for aero-engines, which addresses the requirement for a highly linear mapping between throttle lever and thrust, the construction of a first intermediate / idle thrust estimator includes:

[0021] S201: Within the full envelope, randomly run the engine component-level model to slow and intermediate states;

[0022] S202: Record H, M, T t2 p t2 Record the idle thrust F as the input parameter of the first intermediate / idle thrust estimator. Idl Intermediate thrust F Int As the output parameter of the first intermediate / slow train thrust estimator;

[0023] S203: Divide the acquired data into a training set and a test set;

[0024] S204: On the acquired training set, train the first intermediate / slow train thrust estimator based on a multilayer perceptron neural network, and test the trained first intermediate / slow train thrust estimator with a test set. If the test passes, the training of the first intermediate / slow train thrust estimator is complete.

[0025] In the intelligent thrust control program for aero-engines that addresses the requirement of high linear mapping between throttle lever and thrust, in step S300, the decay estimator includes:

[0026] A baseline parameter estimator is used to estimate baseline values ​​for parameters characterizing the decline.

[0027] A thrust deviation estimator is used to estimate the thrust deviation of the first estimate of intermediate thrust and the thrust deviation of the first estimate of idle thrust based on the deviation of the decay characterization parameters.

[0028] In the intelligent thrust control program for aero-engines that addresses the requirement of high linear mapping between throttle lever and thrust, the reference parameter estimator is expressed as:

[0029] ,

[0030] in, This represents the input-output mapping relationship of the benchmark parameter estimator. and These represent the input and output vectors of the baseline parameter estimator, respectively. , and n L, n H and T t6 The conversion baseline value without considering engine performance degradation; the subscript cor represents the conversion parameter, n L,cor n H,cor and T t6,cor For n L, n H and T t6 The conversion parameters.

[0031] S Eng The engine status is as follows:

[0032] ,

[0033] Among them, F Int,0 and F Idl,0 The reference values ​​for intermediate and idle thrust estimated by the first intermediate / idle thrust estimator; The result is the real-time thrust estimation of the engine, which is a known quantity that can be obtained from existing technologies.

[0034] In the intelligent thrust control program for aero-engines, which addresses the requirement for a highly linear mapping between throttle lever and thrust, the thrust deviation estimator is expressed as:

[0035] ,

[0036] in, This represents the input-output mapping relationship of the thrust deviation estimator. and Let Δn represent the input and output vectors of the thrust deviation estimator, respectively. L,cor , Δn H,cor ΔT t6,cor n L,cor n H,cor T t6,cor The deviation between the measured value and its reference value; and The thrust deviation of the first estimated value of intermediate thrust and the thrust deviation of the first estimated value of idle thrust due to engine performance degradation are the results of the intermediate and idle thrust.

[0037] In the intelligent thrust control program for aero-engines, which addresses the requirement for a highly linear mapping between throttle lever and thrust, the construction of a decay estimator includes,

[0038] S301: Within the full envelope, randomly run the engine component-level model to intermediate states of 50%, 60%, 70%, and 80%;

[0039] S302: Record H, M, T t2 p t2 S Eng As input parameters to the benchmark parameter estimator, record n L,cor n H,cor T t6,cor As the output parameter of the benchmark parameter estimator;

[0040] S303: Divide the data used to build the baseline parameter estimator into a training set and a test set;

[0041] S304: On the acquired training set, train a benchmark parameter estimator based on a multilayer perceptron neural network, and test the trained benchmark parameter estimator with a test set. If the test passes, the benchmark parameter estimator training is complete.

[0042] S305: The efficiency and flow parameters of the fan, high-pressure compressor, high-pressure turbine and low-pressure turbine of the engine component model are biased. Within the full envelope, the engine component model is first randomly run to the intermediate state of 50%, 60%, 70% and 80%, and then randomly run to the slow and intermediate states.

[0043] S306: For parameter n L,cor n H,cor T t6,cor F Idl F Int Record the parameter deviation Δn after the model is pulled. L,cor , Δn H,cor ΔT t6,cor ΔFIdl ΔF Int ;

[0044] S307: Records H, M, T t2 p t2 S Eng , Δn L,cor , Δn H,cor ΔT t6,cor As input parameters for the thrust deviation estimator, ΔF is recorded. Idl ΔF Int As the output parameter of the thrust deviation estimator;

[0045] S308: Divide the data used to build the thrust deviation estimator into a training set and a test set;

[0046] S309: On the acquired training set, a thrust deviation estimator is trained based on a multilayer perceptron neural network, and the trained thrust deviation estimator is tested with a test set. If the test passes, the thrust deviation estimator training is complete.

[0047] In the intelligent thrust control program for aero-engines that addresses the requirement of high linear mapping between throttle lever and thrust, the online update of the second intermediate / idle thrust estimator follows the online update rules of a multilayer perceptron neural network.

[0048] In the intelligent thrust control program for aero-engines that addresses the requirement of high linear mapping between throttle lever and thrust, the reference thrust estimator is:

[0049] ,

[0050] in: Represents a linear interpolation function, PLA Idl and PLA Int These are the throttle lever angles corresponding to slow and intermediate states, respectively. This is the reference thrust estimate result determined by the reference thrust estimator.

[0051] Compared with the prior art, the beneficial effects of this invention are as follows:

[0052] This invention constructs a degradation estimation architecture consisting of a baseline parameter estimator and a thrust deviation estimator. Based on an artificial intelligence neural network, it establishes a nonlinear mapping between performance degradation and intermediate and idle thrust deviations, enabling adaptive correction of the thrust control plan in response to performance degradation. Ultimately, throughout the entire lifespan, it effectively reduces the non-preset dead zone of thrust changes and improves the linearity between the throttle lever and thrust.

[0053] The description provided is merely an overview of the technical solution of this invention. In order to make the technical means of this invention clearer and more understandable, so that those skilled in the art can implement it according to the contents of the specification, and to make the described and other objects, features and advantages of this invention more obvious and understandable, specific embodiments of this invention are described below. Attached Figure Description

[0054] Various other advantages and benefits of the present invention will become apparent to those skilled in the art upon reading the detailed description of the preferred embodiments below. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. It is obvious that the drawings described below are merely some embodiments of the invention, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort. Furthermore, the same reference numerals denote the same parts throughout the drawings.

[0055] In the attached diagram:

[0056] Figure 1 This is a target operating point distribution map provided in one embodiment of the present disclosure for constructing a dataset for intelligent thrust control planning;

[0057] Figure 2 This is a design architecture diagram of an intelligent thrust control scheme for an aero-engine that addresses the requirement for high linear mapping between throttle lever and thrust, provided by another embodiment of this disclosure;

[0058] Figure 3 This is a design architecture diagram of a decay estimator provided in another embodiment of this disclosure;

[0059] Figure 4(a) is a comparison diagram of the relationship between thrust and throttle lever angle during the first 25 hours of engine operation provided in another embodiment of this disclosure; Figure 4(b) is a comparison diagram of the relationship between thrust and throttle lever angle during the first 50 hours of engine operation provided in another embodiment of this disclosure.

[0060] Figure 5(a) shows the thrust variation curve over time under the intelligent thrust control plan provided in another embodiment of this disclosure, with envelope points H=9821.19m and M=0.33, and an operating time of 25 hours; Figure 5(b) shows the thrust variation curve over time under the intelligent thrust control plan provided in another embodiment of this disclosure, with envelope points H=15644.12m and M=1.39, and an operating time of 50 hours.

[0061] The present invention will be further explained below with reference to the accompanying drawings and embodiments. Detailed Implementation

[0062] Specific embodiments of the invention will now be described in more detail with reference to the accompanying drawings. While specific embodiments of the invention are shown in the drawings, it should be understood that the invention can be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided to enable a more thorough understanding of the invention and to fully convey the scope of the invention to those skilled in the art.

[0063] It should be noted that certain terms are used in the specification and claims to refer to specific components. Those skilled in the art will understand that different terms may be used to refer to the same component. This specification and claims do not distinguish components based on differences in terminology, but rather on differences in function. The terms "comprising" or "including" used throughout the specification and claims are open-ended and should be interpreted as "comprising but not limited to." The following descriptions are preferred embodiments for carrying out the invention; however, these descriptions are for the purpose of understanding the general principles of the specification and are not intended to limit the scope of the invention. The scope of protection of this invention is determined by the appended claims.

[0064] To facilitate understanding of the embodiments of the present invention, the following will provide further explanation and description with reference to the accompanying drawings and several specific embodiments, and the accompanying drawings do not constitute a limitation on the embodiments of the present invention.

[0065] To better understand, such as Figure 1 As shown in Figure 5(b), where, as Figure 1 As shown, 10,000 target operating points were randomly generated, each with different altitudes, Mach numbers, and non-standard ambient temperature drift; further details can be found in [link to documentation]. Figure 2 This invention discloses an intelligent thrust control scheme for aero-engines that addresses the requirement of high linear mapping between throttle lever and thrust, comprising the following steps:

[0066] S100: Steady-state operating data is obtained by running an engine component-level model, the steady-state operating data including flight parameters and engine air path parameters;

[0067] S200: Construct a first intermediate / idle thrust estimator to provide a first estimate of intermediate thrust and a first estimate of idle thrust based on flight parameters;

[0068] S300: Construct a decay estimator to estimate the thrust deviation caused by engine performance degradation based on flight parameters and engine air path parameters. The thrust deviation includes the thrust deviation of the first estimated intermediate thrust and the thrust deviation of the first estimated idle thrust.

[0069] S400: Sum the first estimate of intermediate thrust and the thrust deviation of the first estimate of intermediate thrust to calculate the corrected estimate of intermediate thrust; Sum the first estimate of idle thrust and the thrust deviation of the first estimate of idle thrust to calculate the corrected estimate of idle thrust.

[0070] S500: Construct a second intermediate / idle thrust estimator with the same structure as the first intermediate / idle thrust estimator. The second intermediate / idle thrust estimator is used to update online based on the corrected estimate of intermediate thrust and the corrected estimate of idle thrust, and finally provides a second estimate of intermediate thrust and a second estimate of idle thrust based on flight parameters.

[0071] S600: Construct a reference thrust estimator, using the second estimate of intermediate thrust and the second estimate of idle thrust, and determine the reference thrust by linear interpolation of the throttle lever angle.

[0072] It is understood that this invention constructs a degradation estimation architecture consisting of a baseline parameter estimator and a thrust deviation estimator. Based on an artificial intelligence neural network, it establishes a nonlinear mapping between performance degradation and intermediate and idle thrust deviations, achieving adaptive correction of the thrust control plan in response to performance degradation. This means that the intelligent thrust control plan can replace the traditional speed control plan. Throughout its entire lifespan, this invention can effectively reduce the non-preset dead zone of thrust changes and improve the linearity between the throttle lever and thrust.

[0073] In a preferred embodiment of the intelligent thrust control scheme for aero-engines that addresses the requirement for a high linear mapping between throttle lever and thrust, flight parameters include altitude H, Mach number M, and throttle lever PLA; engine air path parameters include low-pressure rotor speed n. L High-voltage rotor speed n H Total temperature T at the fan inlet t2 Low-pressure turbine outlet total temperature T t6 Total pressure at the fan inlet p t2 .

[0074] In a preferred embodiment of the intelligent thrust control scheme for aero-engines that addresses the requirement of high linear mapping between throttle lever and thrust, the first intermediate / idle thrust estimator is:

[0075] ,

[0076] in, This represents the input-output mapping relationship of the first intermediate / slow train thrust estimator. and These are the input and output vectors of the first intermediate / slow train thrust estimator, respectively.

[0077] In a preferred embodiment of an intelligent thrust control program for aero-engines that addresses the requirement for a high linear mapping between throttle lever and thrust, constructing a first intermediate / idle thrust estimator includes,

[0078] S201: Within the full envelope, randomly run the engine component-level model to slow and intermediate states;

[0079] S202: Record H, M, T t2 p t2 Record the idle thrust F as the input parameter of the first intermediate / idle thrust estimator. Idl Intermediate thrust F Int As the output parameter of the first intermediate / slow train thrust estimator;

[0080] S203: Divide the acquired data into a training set and a test set;

[0081] S204: On the acquired training set, train the first intermediate / slow train thrust estimator based on a multilayer perceptron neural network, and test the trained first intermediate / slow train thrust estimator with a test set. If the test passes, the training of the first intermediate / slow train thrust estimator is complete.

[0082] The training of a multilayer perceptron neural network can be summarized in three steps: forward propagation, back propagation, and parameter update. The forward propagation process is described as follows:

[0083] ,

[0084] in, and These represent the input and output vectors of the current fully connected layer, respectively. and These represent the weight matrix and the bias vector, respectively.

[0085] Following the fully connected layer is the activation layer, which typically uses the ReLU function, as shown below:

[0086] ,

[0087] in, and These represent the input and output vectors of the activation layer, respectively.

[0088] The loss function is designed based on the mean squared error, as shown below:

[0089] ,

[0090] in, Represents the target vector; This indicates the prediction result.

[0091] Based on the Adam algorithm, the backpropagation and parameter update process is described as follows:

[0092] ,

[0093] in, and These represent the updated weight matrix and bias vector, respectively. and These represent the gradients of the weight matrix and the bias vector, respectively. This represents the learning rate. It should be noted that the online update process of a fully connected neural network still follows the algorithm described above.

[0094] In a preferred embodiment of the intelligent thrust control scheme for aero-engines that addresses the requirement for a high linear mapping between throttle lever and thrust, see [reference needed]. Figure 3 In step S300, the decay estimator includes:

[0095] A baseline parameter estimator is used to estimate baseline values ​​for parameters characterizing the decline.

[0096] A thrust deviation estimator is used to estimate the thrust deviation of the first estimate of intermediate thrust and the thrust deviation of the first estimate of idle thrust based on the deviation of the decay characterization parameters.

[0097] In a preferred embodiment of an aero-engine intelligent thrust control program addressing the requirement for a high linear mapping between throttle lever and thrust, the reference parameter estimator is expressed as:

[0098] ,

[0099] in, This represents the input-output mapping relationship of the benchmark parameter estimator. and These represent the input and output vectors of the baseline parameter estimator, respectively. , and n L, n H and T t6 The conversion benchmark value without considering engine performance degradation.

[0100] S Eng The engine status is as follows:

[0101] ;

[0102] Among them, F Int,0 and F Idl,0 The reference values ​​for intermediate and idle thrust estimated by the first intermediate / idle thrust estimator; The result is the real-time thrust estimation of the engine, which is a known quantity that can be obtained from existing technologies.

[0103] In a preferred embodiment of an aero-engine intelligent thrust control program that addresses the requirement for a highly linear mapping between throttle lever and thrust, the thrust deviation estimator is expressed as:

[0104] ,

[0105] in, This represents the input-output mapping relationship of the thrust deviation estimator. and Let Δn represent the input and output vectors of the thrust deviation estimator, respectively. L,cor , Δn H,cor ΔT t6,cor n L,cor n H,cor T t6,cor The deviation between the measured value and its reference value; and The thrust deviation of the first estimated value of intermediate thrust and the thrust deviation of the first estimated value of idle thrust due to engine performance degradation are the results of the intermediate and idle thrust.

[0106] In a preferred embodiment of an aero-engine intelligent thrust control program addressing the requirement for a high linear mapping between throttle lever and thrust, constructing a decay estimator includes,

[0107] S301: Within the full envelope, randomly run the engine component-level model to intermediate states of 50%, 60%, 70%, and 80%;

[0108] S302: Record H, M, T t2 p t2 S Eng As input parameters to the benchmark parameter estimator, record n L,cor n H,cor T t6,cor As the output parameter of the benchmark parameter estimator;

[0109] S303: Divide the data used to build the baseline parameter estimator into a training set and a test set;

[0110] S304: On the acquired training set, train a benchmark parameter estimator based on a multilayer perceptron neural network, and test the trained benchmark parameter estimator with a test set. If the test passes, the benchmark parameter estimator training is complete.

[0111] S305: The efficiency and flow parameters of the fan, high-pressure compressor, high-pressure turbine and low-pressure turbine of the engine component model are biased. Within the full envelope, the engine component model is first randomly run to the intermediate state of 50%, 60%, 70% and 80%, and then randomly run to the slow and intermediate states.

[0112] S306: For parameter n L,cor n H,cor T t6,cor F Idl F Int Record the parameter deviation Δn after the model is pulled. L,cor , Δn H,cor ΔT t6,cor ΔF Idl ΔF Int ;

[0113] S307: Records H, M, T t2 p t2 S Eng , Δn L,cor , Δn H,cor ΔT t6,cor As input parameters for the thrust deviation estimator, ΔF is recorded. Idl ΔF Int As the output parameter of the thrust deviation estimator;

[0114] S308: Divide the data used to build the thrust deviation estimator into a training set and a test set;

[0115] S309: On the acquired training set, a thrust deviation estimator is trained based on a multilayer perceptron neural network, and the trained thrust deviation estimator is tested with a test set. If the test passes, the thrust deviation estimator training is complete.

[0116] In a preferred embodiment of the intelligent thrust control scheme for aero-engines that addresses the requirement for a high linear mapping between throttle lever and thrust, the online update of the second intermediate / idle thrust estimator follows the online update rules of a multilayer perceptron neural network.

[0117] In a preferred embodiment of an aero-engine intelligent thrust control scheme requiring a high linear mapping between throttle lever and thrust, the reference thrust estimator is:

[0118] ,

[0119] in: Represents a linear interpolation function, PLA Idl and PLA Int These are the throttle lever angles corresponding to slow and intermediate states, respectively. This is the reference thrust estimate result determined by the reference thrust estimator.

[0120] In one embodiment, a first intermediate / idle thrust estimator is responsible for providing a first estimate of intermediate thrust and a first estimate of idle thrust; a degradation estimator is responsible for estimating and compensating for the deviation of intermediate and idle thrust caused by engine performance degradation; a second intermediate / idle thrust estimator is updated online based on the compensated estimates of intermediate and idle thrust to adapt to continuous engine performance degradation; and a reference thrust estimator determines the final reference thrust by linear interpolation based on the output of the second intermediate / idle thrust estimator and the current throttle lever angle.

[0121] The recession estimator involves recession characterization parameters n. L n H T t6 The conversion parameter is denoted as n. L,cor n H,cor T t6,cor .

[0122] As shown in Figures 4(a) to 4(b), compared with the traditional speed control scheme, the intelligent thrust control scheme proposed in this invention significantly reduces the non-preset dead zone range of thrust change and improves the linearity of the correspondence between thrust and throttle lever angle.

[0123] As shown in Figures 5(a) and 5(b), two envelope points were selected to illustrate the thrust-time variation curves under the intelligent thrust control scheme. At the second envelope point, a 2° non-preset dead zone was observed near the engine's intermediate state. This fully demonstrates that the intelligent thrust control scheme proposed in this invention has significant advantages in improving the correspondence between thrust and throttle lever angle, and can effectively reduce the pilot's operational burden on the throttle lever angle.

[0124] In another embodiment, the first and second intermediate / idle thrust estimators and the decay estimator constructed in this invention are all built using a multilayer perceptron (MLP) neural network. This is because: the inventors recognize that MLPs possess powerful nonlinear mapping capabilities, are particularly suitable for modeling complex aerodynamic and thermodynamic relationships of engines, support multi-input multi-output modeling, and are especially adaptable to multi-dimensional inputs of flight parameters and engine states; after training, the MLP model can achieve high-precision thrust estimation and deviation prediction; and through continuous optimization of the model via an online learning mechanism, it can adapt to changes in engine performance, thereby improving the intelligence and adaptability of the control plan. Specifically, based on altitude H, Mach number M, and fan inlet total temperature T... t2 and fan inlet total pressure p t2As input, the first intermediate / idle thrust estimator provides a baseline estimate of the intermediate thrust and a first estimate of the idle thrust. The decay estimator estimates the thrust deviation caused by engine performance degradation and compensates for the baseline estimate (i.e., adding the corresponding first estimate and the thrust deviation mentioned above) to obtain corrected estimates of the intermediate and idle thrust respectively. The second intermediate / idle thrust estimator updates online based on the corrected estimates of the intermediate and idle thrust to adapt to engine performance degradation, obtaining decay-adaptive intermediate and idle thrust estimates. Subsequently, the reference thrust estimator constructed in this invention uses the output of the second intermediate / idle thrust estimator to determine the reference thrust by linearly interpolating the throttle lever angle. The decay estimator consists of a baseline parameter estimator and a thrust deviation estimator. The baseline parameter estimator is based on height H, Mach number M, and fan inlet total temperature T. t2 Total pressure at the fan inlet p t2 and engine status S Eng As input, estimate the fan's equivalent speed n. L,cor Calculation of high-pressure compressor speed n H,cor After the low-pressure turbine, the total temperature T is converted. t6,cor The baseline values ​​for these three degradation characterization parameters; the thrust deviation estimator is based on altitude H, Mach number M, and total fan inlet temperature T. t2 Total pressure at the fan inlet p t2 Engine Status S Eng Fan conversion speed deviation Δn L,cor Δn, the converted speed deviation of the high-pressure compressor H,cor The total temperature deviation ΔT after the low-pressure turbine is calculated. t6,cor As input, estimate the intermediate thrust deviation ΔF Int and the deviation of the slow train thrust ΔF Idl This is the output of the recession estimator.

[0125] The basic principles of this application have been described above with reference to specific embodiments. However, it should be noted that the advantages, benefits, and effects mentioned in this application are merely examples and not limitations, and should not be considered as essential features of each embodiment of this application. Furthermore, the specific details disclosed above are for illustrative and facilitative purposes only, and are not limitations. These details do not limit the application to the necessity of employing the aforementioned specific details for implementation.

[0126] The above description has been given for purposes of illustration and description. Furthermore, this description is not intended to limit the embodiments of this application to the forms disclosed herein. Although numerous exemplary aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, alterations, additions, and sub-combinations thereof.

Claims

1. A smart thrust control scheme for aero-engines that addresses the requirement of high linear mapping between throttle lever and thrust, characterized in that, It includes the following steps, S100: Steady-state operating data is obtained by running an engine component-level model, the steady-state operating data including flight parameters and engine air path parameters; S200: Construct a first intermediate / idle thrust estimator to provide a first estimate of intermediate thrust and a first estimate of idle thrust based on flight parameters; S300: Construct a decay estimator to estimate the thrust deviation caused by engine performance degradation based on flight parameters and engine air path parameters. The thrust deviation includes the thrust deviation of the first estimated intermediate thrust and the thrust deviation of the first estimated idle thrust. S400: Sum the first estimate of intermediate thrust and the thrust deviation of the first estimate of intermediate thrust to calculate the corrected estimate of intermediate thrust; Sum the first estimate of idle thrust and the thrust deviation of the first estimate of idle thrust to calculate the corrected estimate of idle thrust. S500: Construct a second intermediate / idle thrust estimator with the same structure as the first intermediate / idle thrust estimator. The second intermediate / idle thrust estimator is used to update online based on the corrected estimate of intermediate thrust and the corrected estimate of idle thrust, and finally provides a second estimate of intermediate thrust and a second estimate of idle thrust based on flight parameters. S600: Construct a reference thrust estimator, using the second estimate of intermediate thrust and the second estimate of idle thrust, and determine the reference thrust by linear interpolation of the throttle lever angle.

2. The intelligent thrust control scheme for aero-engines according to claim 1, which addresses the requirement of high linear mapping between throttle lever and thrust, is characterized in that... Preferably, the flight parameters include altitude H, Mach number M, and throttle lever PLA; the engine air path parameters include low-pressure rotor speed n. L High-voltage rotor speed n H Total temperature T at the fan inlet t2 Low-pressure turbine outlet total temperature T t6 Total pressure at the fan inlet p t2 .

3. The intelligent thrust control scheme for aero-engines according to claim 1, which addresses the requirement for high linear mapping between throttle lever and thrust, is characterized in that... The first intermediate / slow train thrust estimator is: ， in, This represents the input-output mapping relationship of the first intermediate / slow train thrust estimator. and These are the input and output vectors of the first intermediate / slow train thrust estimator, respectively.

4. The intelligent thrust control scheme for aero-engines according to claim 1, which addresses the requirement for high linear mapping between throttle lever and thrust, is characterized in that... Constructing the first intermediate / slow train thrust estimator includes, S201: Within the full envelope, randomly run the engine component-level model to slow and intermediate states; S202: Record H, M, T t2 p t2 Record the idle thrust F as the input parameter of the first intermediate / idle thrust estimator. Idl Intermediate thrust F Int As the output parameter of the first intermediate / slow train thrust estimator; S203: Divide the acquired data into a training set and a test set; S204: On the acquired training set, train the first intermediate / slow train thrust estimator based on a multilayer perceptron neural network, and test the trained first intermediate / slow train thrust estimator with a test set. If the test passes, the training of the first intermediate / slow train thrust estimator is complete.

5. The intelligent thrust control scheme for aero-engines according to claim 1, which addresses the requirement of high linear mapping between throttle lever and thrust, is characterized in that... In step S300, the decay estimator includes: A baseline parameter estimator is used to estimate baseline values ​​for parameters characterizing the decline. A thrust deviation estimator is used to estimate the thrust deviation of the first estimate of intermediate thrust and the thrust deviation of the first estimate of idle thrust based on the deviation of the decay characterization parameters.

6. The intelligent thrust control scheme for aero-engines according to claim 5, which addresses the requirement for high linear mapping between throttle lever and thrust, is characterized in that... The baseline parameter estimator is represented as follows: ， in, This represents the input-output mapping relationship of the benchmark parameter estimator. and These represent the input and output vectors of the baseline parameter estimator, respectively. , and n L, n H and T t6 The conversion baseline value without considering engine performance degradation; the subscript cor represents the conversion parameter, n L,cor n H,cor and T t6,cor For n L, n H and T t6 Conversion parameters; S Eng The engine status is as follows: ， Among them, F Int,0 and F Idl,0 The reference values ​​for intermediate and idle thrust estimated by the first intermediate / idle thrust estimator; This is the real-time thrust estimation result for the engine, and it represents a known quantity.

7. The intelligent thrust control scheme for aero-engines according to claim 5, which addresses the requirement of high linear mapping between throttle lever and thrust, is characterized in that... The thrust deviation estimator is expressed as: ， in, This represents the input-output mapping relationship of the thrust deviation estimator. and Let Δn represent the input and output vectors of the thrust deviation estimator, respectively. L,cor , Δn H,cor ΔT t6,cor n L,cor n H,cor T t6,cor The deviation between the measured value and its reference value; and The thrust deviation of the first estimated value of intermediate thrust and the thrust deviation of the first estimated value of idle thrust due to engine performance degradation are the results of the intermediate and idle thrust.

8. The intelligent thrust control scheme for aero-engines according to claim 1, which addresses the requirement for high linear mapping between throttle lever and thrust, is characterized in that... Constructing a recession estimator includes, S301: Within the full envelope, randomly run the engine component-level model to intermediate states of 50%, 60%, 70%, and 80%; S302: Record H, M, T t2 p t2 S Eng As input parameters to the benchmark parameter estimator, record n L,cor n H,cor T t6,cor As the output parameter of the benchmark parameter estimator; S303: Divide the data used to build the baseline parameter estimator into a training set and a test set; S304: On the acquired training set, train a benchmark parameter estimator based on a multilayer perceptron neural network, and test the trained benchmark parameter estimator with a test set. If the test passes, the benchmark parameter estimator training is complete. S305: The efficiency and flow parameters of the fan, high-pressure compressor, high-pressure turbine and low-pressure turbine of the engine component model are biased. Within the full envelope, the engine component model is first randomly run to the intermediate state of 50%, 60%, 70% and 80%, and then randomly run to the slow and intermediate states. S306: For parameter n L,cor n H,cor T t6,cor F Idl F Int Record the parameter deviation Δn after the model is pulled. L,cor , Δn H,cor ΔT t6,cor ΔF Idl ΔF Int ; S307: Records H, M, T t2 p t2 S Eng , Δn L,cor , Δn H,cor ΔT t6,cor As input parameters for the thrust deviation estimator, ΔF is recorded. Idl ΔF Int As the output parameter of the thrust deviation estimator; S308: Divide the data used to build the thrust deviation estimator into a training set and a test set; S309: On the acquired training set, a thrust deviation estimator is trained based on a multilayer perceptron neural network, and the trained thrust deviation estimator is tested with a test set. If the test passes, the thrust deviation estimator training is complete.

9. The intelligent thrust control scheme for aero-engines according to claim 1, which addresses the requirement for high linear mapping between throttle lever and thrust, is characterized in that... The online update of the second intermediate / slow train thrust estimator follows the online update rules of the multilayer perceptron neural network.

10. The intelligent thrust control scheme for aero-engines according to claim 1, which addresses the requirement of high linear mapping between throttle lever and thrust, is characterized in that... The reference thrust estimator is: ， in: Represents a linear interpolation function, PLA Idl and PLA Int These are the throttle lever angles corresponding to slow and intermediate states, respectively. This is the reference thrust estimate result determined by the reference thrust estimator.

Images